The emergence of serum antibody responses only after early cellular immune responses have suppressed HIV replication has led to doubts regarding the importance of humoral immunity in controlling HIV infections. Nonetheless, rare, broadly cross-reactive antibodies are capable of neutralizing multiple isolates of HIV-1 in vitro, and when passively administered to monkeys, prevent experimental infection by SHIV. Why are such antibodies rarely produced by HIV- infected patients? Recently, several of these rare, HIV neutralizing antibodies were shown to react with self-antigens leading to the possibility that effective HIV humoral responses are constrained by the tolerization of HIV-reactive B cells that also recognize self-antigens. We shall test the hypothesis that B cells that recognize phylogenetically conserved, neutralizing epitopes of the HIV-1 gp-41 membrane proximal external region (MPER) are present in early, developmentally immature B cells but are tolerized and lost during their subsequent development/maturation.